Nofollow Links SEO: rel=nofollow, sponsored, and ugc Explained

The rel="nofollow" link attribute was introduced by Google in 2005 as a way to tell crawlers not to follow a link or pass PageRank through it. In 2019, Google changed how it treats nofollow: it became a hint rather than a directive, meaning Google may choose to crawl and index nofollow links despite the attribute. It also introduced two new attributes — rel="ugc" and rel="sponsored" — to give webmasters more specific ways to categorise link types.

The three link relationship attributes

Since September 2019, there are three link relationship attributes for telling Google something about a link's nature:

rel="nofollow" — the original catch-all attribute. Tells Google you don't want to vouch for the link. Used when none of the more specific attributes apply: links in untrusted areas, links added by users, links you don't endorse, links to pages you don't want to be associated with.

rel="ugc" — User-Generated Content. For links created by users in comments, forum posts, wiki contributions, reviews, and other user-generated areas of a site. Signals to Google that the link was placed by a third party and not by the site owner.

rel="sponsored" — For paid links: sponsored posts, paid placements, paid reviews, affiliate links, advertisements. Required by Google's guidelines for any link where money has changed hands or the link is part of an advertising arrangement.

These attributes can be combined: rel="ugc nofollow" is valid and used by sites that want to both mark content as user-generated and flag it as nofollow.

What changed in 2019: hint vs directive

Before 2019, Google treated rel="nofollow" as a directive — a firm instruction not to count the link for PageRank purposes. Google did still crawl nofollow links (the attribute doesn't prevent crawling by itself), but the link's PageRank signal was discarded.

In September 2019, Google announced that it would treat all three link attributes — nofollow, ugc, and sponsored — as hints rather than directives. This means Google may use these links for ranking purposes when it deems appropriate, ignoring the attribute's intent. Google's stated reason: it uses all available signals to understand the web, and nofollow hints are factored into its crawling and indexing decisions rather than being absolute.

Practically, this means:

• Nofollow links may pass some PageRank in some circumstances
• Google may still choose to crawl and index nofollow links
• The attribute is still a useful signal to Google about the nature of the link
• Nofollowing outbound links to editorial content you don't endorse is still correct practice
• You cannot reliably use nofollow to sculpt PageRank within your own site

Bing continued to treat nofollow as a directive (not following nofollowed links for PageRank) — so nofollow still has its original effect on Bing.

When to use rel="sponsored"

Use rel="sponsored" on any link where value has been exchanged for the link:

• Affiliate links (you earn commission when a user clicks or purchases)
• Sponsored posts and paid content placements
• Paid review links
• Links in paid advertisements within editorial content
• Any link that exists because of a commercial arrangement

Google's link spam guidelines require sponsored designation. Failure to mark paid links as rel="sponsored" (or at minimum rel="nofollow") is a violation of Google's guidelines and the type of link pattern that can trigger a manual action for unnatural links. The risk is primarily on the linking site (the one hosting the paid link), not the receiving site, though receiving sites with many paid links pointing to them may face algorithmic scrutiny.

When to use rel="ugc"

Use rel="ugc" on links created by users in areas of your site where you don't editorially control the content:

• Blog comment sections
• Forum posts
• Community Q&A threads
• Wiki contributions by unverified users
• User review links
• Any area where arbitrary users can add links without editorial approval

The practical effect is similar to nofollow — it signals to Google that you, the site owner, did not add or endorse this link. Many CMS plugins (WordPress comment forms, Discourse, phpBB) automatically add rel="nofollow ugc" to user-contributed links. You can use either rel="nofollow" or rel="ugc" for this purpose; Google treats them similarly, but ugc is more semantically specific.

When to use rel="nofollow" on outbound links

For links you've added yourself (not user-generated, not paid), use rel="nofollow" when:

• Linking to a site you're citing but don't endorse (e.g., linking to a low-quality site to call it out)
• Linking to a competitor's site in a way you don't want to interpret as a vote
• Linking to login pages, terms of service, privacy pages, or other utility pages where passing PageRank is irrelevant
• Linking to untrusted external resources where you're uncertain of quality

You don't need to nofollow normal editorial outbound links to high-quality sources. Google's guidelines encourage linking to relevant resources — an outbound link to a reputable authoritative source on the topic you're discussing is natural and appropriate. Nofollowing all outbound links out of habit is unnecessary and can actually look unnatural.

The internal nofollow mistake

One of the most common technical SEO errors is applying rel="nofollow" to internal links — links within your own site from one page to another. This is almost never appropriate and actively harms your internal PageRank distribution.

Nofollow on internal links:

• Prevents Google from following the link (reducing the chance the destination page is crawled via this path)
• Wastes the source page's PageRank — the PageRank that would have flowed through this link is lost, it doesn't get redistributed to other links on the page
• Doesn't hide the destination page from Google (if the page is linked elsewhere, Google still finds and indexes it)

The original motivation for "PageRank sculpting" (nofollowing some internal links to concentrate PageRank on others) was invalidated in 2009 when Google clarified that nofollow links don't cause PageRank to flow to other links — the PageRank is simply lost. Post-2019's hint treatment makes internal nofollowing even less useful.

The correct approach for internal links you don't want crawled or indexed: if the page should not be indexed, add <meta name="robots" content="noindex"> to the destination page. If it should not be crawled, add it to robots.txt. Neither of these requires nofollowing the links pointing to it.

Nofollow and affiliate links: the correct approach

Affiliate links should always carry rel="sponsored" (or at minimum rel="nofollow"). Affiliate links with no attribute are effectively undisclosed paid links — a violation of Google's link spam guidelines. The risk:

• If Google identifies your site as hosting many unmarked affiliate links (links to affiliate networks, links with affiliate tracking parameters), it may issue a manual action for "unnatural links"
• Sites with heavy unmarked affiliate content are frequently targeted by Helpful Content System classifiers

Correct affiliate link markup:

<a href="https://example.com/product?affiliate_id=123" rel="sponsored nofollow">Product Name</a>

Using both sponsored and nofollow is belt-and-suspenders — sponsored is the semantically correct attribute; adding nofollow ensures older crawlers that don't understand sponsored also handle the link correctly.

How to audit your site's nofollow usage

A thorough nofollow audit checks for three issues:

1. Unmarked paid links. Search your content for affiliate tracking parameters (patterns like ?ref=, ?aff_id=, ?partner=, known affiliate network domains like shareasale.com, impact.com, cj.com) and check whether those links carry rel="sponsored" or rel="nofollow". Missing designation on known affiliate links is the highest-risk finding.

2. Internal nofollows. Crawl your site and check for rel="nofollow" on links where both the source and destination have your domain. These should be removed unless there's a deliberate reason.

3. Over-aggressive nofollow on editorial links. Review whether you're nofollowing legitimate outbound editorial links to authoritative sources. This is harmless from a penalty perspective but wastes any potential signals from your outbound linking context.

FAQ

Do nofollow links have any SEO value?

After Google's 2019 hint change, nofollow links may have some SEO value — Google may choose to use them for ranking purposes as hints. Before 2019, the conventional wisdom was "nofollow links pass no PageRank." The current reality is more nuanced: nofollow links from high-authority sources may contribute something, but you cannot rely on it. For practical purposes: focus on earning dofollow editorial links (the clearest positive signal); don't specifically pursue nofollow links from directories or low-quality sources expecting PageRank benefit.

Should I nofollow all outbound links to protect my PageRank?

No — this is a misconception. Nofollowing editorial outbound links doesn't "save" your PageRank by keeping it on your site. Post-2009, Google confirmed that PageRank "lost" to a nofollow link is simply gone — it doesn't redistribute to other outbound links on the page. Nofollowing all outbound links also looks unnatural and removes a contextual relevance signal that Google uses to understand what your page is about. Add nofollow only when it's appropriate for the specific link type — don't use it as a PageRank conservation strategy.

What's the difference between nofollow and disavow?

rel="nofollow" is placed by the linking site on links it's placing on its own pages. Disavow is used by the receiving site to tell Google to ignore specific inbound links pointing to it. If you have toxic links pointing to your site, you use the disavow tool (in Google Search Console) — you cannot add nofollow to links on other sites. Nofollow is used when you control the page with the link; disavow is used when you don't control the linking page.

Does rel="nofollow" prevent a page from being indexed?

No. rel="nofollow" prevents the link from passing PageRank and (before 2019) prevented Google from following that specific link path. It does not prevent the destination page from being indexed if Google finds it through any other means (other links, sitemap, direct fetch). To prevent indexing, use <meta name="robots" content="noindex"> on the destination page itself, or block it in robots.txt.